Auto Generated Shorts Videos using Manim and ChatTTS

This project show how to generate a short video without any editing or recording. Everything is completed by pure coding and model generation.

Demo (Turn on the volume icon - by default its mute)

final_video.mp4

Short Youtube: https://www.youtube.com/shorts/rv0BuhHG9TQ

Free free to give star ⭐ for this project.

Pre-requisites

I'm using Ubuntu 24.04, Miniforge Python 3.11. Packages needed are :

1. Manim for video creation ( https://github.com/ManimCommunity/manim )
2. ChatTTS for audio generation ( https://github.com/2noise/ChatTTS )
3. FFMPEG for audio + video merging

Please follow each step installation.

Setup and Generate the video

git clone https://github.com/yodiaditya/manim-shorts
cd manim-shorts
manim -pqk main.py Main

Generate audio

The pre-generated audio located at assets and for voice embedding in embedding folder.

If you want to know, please check audio.py

python audio.py

Merging Audio and Video

Copy the output of Manim video into assets folder. Then go inside and run this

ffmpeg -i Main.mp4 -i 1.wav -i 2.wav -i 3.wav -i silent.wav -i 4.wav -filter_complex \
"[1:a][2:a]concat=n=2:v=0:a=1[a1]; \
 [a1][3:a]concat=n=2:v=0:a=1[a2]; \
 [a2][4:a]concat=n=2:v=0:a=1[a3]; \
 [a3][5:a]concat=n=2:v=0:a=1[aout]" \
-map 0:v -map "[aout]" -c:v copy -c:a aac final_video.mp4

Credits

By @yodiaditya. Happy to connect with you over Linkedin and Github!